Fe3O4 Nanoparticles Anchored on Carbon Serve the Dual Role of Catalyst and Magnetically Recoverable Entity in the Aerobic Oxidation of Alcohols

Article abstract of DOI:10.1002/cctc.201501149

A composite of Fe₃O₄ nanoparticles anchored on a carbon support (Fe₃O₄/C), possessing both superparamagnetism and molecular oxygen activating properties, was prepared by an ammonia-assisted precipitation method. Fe₃O₄/C could catalyze the selective oxidation of various benzyl alcohols with air as the oxidant source, and could be easily separated and recycled with an external magnet. The small particle size and the interaction between the Fe₃O₄ nanoparticles and carbon support endow the Fe₃O₄/C catalyst with relatively high reducibility. Its oxidation state is easy to change. This intrinsic property of the Fe₃O₄ nanoparticles could be responsible for the high activity of Fe₃O₄/C in the aerobic oxidation of alcohols.

Full text of DOI:10.1002/cctc.201501149

DOI: 10.1002/cctc.201501149
Full Papers
Fe O Nanoparticles Anchored on Carbon Serve the Dual
3
4
Role of Catalyst and Magnetically Recoverable Entity in
the Aerobic Oxidation of Alcohols
[
a]
[a]
[b]
[a]
[a]
Longlong Geng, Bin Zheng, Xiang Wang, Wenxiang Zhang, Shujie Wu,
[a]
[c]
[a]
Mingjun Jia, Wenfu Yan, and Gang Liu*
A composite of Fe O nanoparticles anchored on a carbon sup-
with an external magnet. The small particle size and the inter-
action between the Fe O nanoparticles and carbon support
3
4
port (Fe O /C), possessing both superparamagnetism and mo-
3
4
3
4
lecular oxygen activating properties, was prepared by an am-
monia-assisted precipitation method. Fe O /C could catalyze
endow the Fe O /C catalyst with relatively high reducibility. Its
3 4
oxidation state is easy to change. This intrinsic property of the
Fe O nanoparticles could be responsible for the high activity
3
4
the selective oxidation of various benzyl alcohols with air as
the oxidant source, and could be easily separated and recycled
3
4
of Fe O /C in the aerobic oxidation of alcohols.
3
4
Introduction
[
16]
Iron is the second most abundant metal in the Earth’s crust.
The advantages of ready availability, low cost, and low toxicity
temperatures (40–808C) and ambient pressure (1 atm). High
concentrations of low-coordinated metal atoms in these quite
small particles led to a high reducibility (facile changing of the
oxidation state) of the iron oxides and, thus, caused relatively
high reduction–oxidation activity. One drawback was that the
amorphous nature gave the ultrafine iron oxides particles very
low superparamagnetism. The catalyst, thus, could not be recy-
cled by separation with an external magnet.
[1–5]
make iron an ideal metal for catalysis.
It is well known that
iron takes part in many biological systems, for example, in
[6,7]
metalloproteins for the transportation of oxygen.
Imitating
such processes, many efficient homogeneous iron catalysts
have been developing for catalyzing cross-coupling reactions,
sulfide oxidations, allylic alkylations and aminations, alcohol ox-
[8–11]
idations and Michael additions, etc.
One desirable charac-
As magnetic entities, magnetic iron oxides (Fe O4 and g-
3
teristic is that most of these iron-based homogeneous catalysts
are suitable for working under relatively mild conditions, that
is, low temperature and ambient pressure.
Fe O , for example) are widely used in heterogeneous catalysis
2
3
as vehicles for supporting catalytically active nanoparticles or
[
17–21]
molecules.
They can simplify the recovery process of cata-
In heterogeneous catalysis, iron-based catalysts (mainly iron
oxides and iron salts) usually work at relatively high tempera-
tures and pressures. Few of them are catalytically active under
lysts from liquid reaction mixtures by using an external super-
magnet. The catalytically active phases mainly consist of orga-
nocatalysts, metal complexes, noble-metal nanoparticles, or en-
[12–14]
[22–27]
mild reaction conditions.
Very recently, we reported that
zymes.
Both Fe O and g-Fe O are poorly or not active in
3 4 2 3
ultrafine iron oxide species (<2 nm) supported on carbon be
aerobic oxidation reactions. Generally, good crystallization is
necessary for magnetic entities with high superparamagnetism.
However, well-defined crystal structures could reduce the sur-
face defects, that is, catalytic active sites, of the material. Even
if forming nanoparticles, crystalline iron oxide could only cata-
lyze the reaction with H O or TBHP (tert-butyl hydroperoxide)
active in the selective oxidation of benzyl alcohol with air as
[
15]
the oxygen source. This catalyst could work at low reaction
[
a] L. Geng, B. Zheng, Prof. W. Zhang, S. Wu, Prof. M. Jia, Prof. G. Liu
Key Laboratory of Surface and Interface Chemistry of Jilin Province
College of Chemistry
2
2
as the oxidants under low temperature and ambient pressure,
[
28–30]
Jilin University
not with atmospheric oxygen.
It seems that there is a gap
Jiefang Road 2519, Changchun, 130012 (P.R. China)
E-mail: lgang@jlu.edu.cn
between the magnetic properties and the catalytic performan-
ces of iron oxide catalysts, especially for activating molecular
[
b] Dr. X. Wang
O under mild conditions.
2
Institute for Integrated Catalysis
Pacific Northwest National Laboratory
Richland, WA 99352 (USA)
In this work, it was found that a composite of carbon-sup-
ported Fe O (denoted as Fe O /C) could serve the dual role of
3
4
3
4
[
c] Prof. W. Yan
catalyst and magnetically recoverable entity in the oxidation
reaction. Fe O /C was prepared by an ammonia-assisted pre-
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry
College of Chemistry
Jilin University
3
4
cipitation method. The obtained Fe O /C possesses relatively
3
4
high superparamagnetism, and can directly catalyze the selec-
tive oxidation of various benzyl alcohols with air as the oxidant
source. It could be easily recovered from liquid reaction mix-
Qianjin Street 2519, Changchun, 130012 (P.R. China)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201501149.
ChemCatChem 2016, 8, 805 – 811
805
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
tures by using an external supermagnet and reused more than
five times. The correlation between the intrinsic properties of
Fe O nanoparticles and their catalytic performance was also
From Figure 1B and the corresponding fitting analysis
(Table 1), we can confirm that the Fe O /C sample is made of
3
4
magnetite (Fe O ). The spectrum can be fitted with two mag-
3
4
3
4
[
37]
investigated based on a series of characterizations.
netic hyperfine sextets. Statistical analysis of the spectrum
showed that the amount of magnetite in the sample is higher
than 86%. There is a broad absorption in the center of the
spectrum, which points to the presence of fluctuation-broad-
Results and Discussion
3
+
ened spectral components. A doublet corresponding to Fe
Fe O /C was prepared by a modified precipitation method
[34]
3
4
species can be found in the spectrum. The Raman spectra of
Fe O /C and bulk Fe O were also measured (Figure 1C). The
(
Scheme 1). Ethanol and ammonia were employed as solvent
3
4
3
4
and alkali precipitator, respectively. A kind of mesoporous
À1
peaks at 1358 and 1598 cm can be assigned to the D and G
bands of the carbon support, respectively (Figure S1b). These
two peaks are very broad, showing the amorphous character
2
À1
carbon with 6.5 nm pore size and 1392 m g surface area was
used as the carbon support. This carbon material was prepared
with a modified hard-template route, which was reported by
À1
of the carbon support. The peaks at 513 and 660 cm can be
[
31,32]
our group previously.
A large amount of oxygen-contain-
[37,38]
assigned to the E and A modes of Fe O .
The relatively
g
1g
3
4
À1
ing functional groups (total amount is about 3.23 mmolg ,
detected by Boehm titration) was present on the surface of
low peak intensity compared with that of bulk Fe O should be
3
4
due to the small particle size of Fe O in Fe O /C. This result
3
4
3
4
À1
the carbon support, including carboxylic (0.92 mmolg ), lac-
further confirms that Fe O4 is the main phase of Fe O /C,
3
3
4
À1
À1 [16]
tonic (0.91 mmolg ), and phenolic groups (1.40 mmolg ).
These functional groups endow the carbon surface with hydro-
philic properties. This facilitated the introduction of iron nitrate
ethanol solution and ammonia aqueous solution into the mes-
opores of the carbon support. In addition, the surface oxygen
functional groups could also serve as strongly active sites for
which is in agreement with the results of the XRD and Mçssba-
uer measurements.
N2 adsorption–desorption isotherms showed that Fe O /C
3
4
possesses a type IV isotherm accompanied by a H3 hysteresis
loop in the relative pressure range 0.4–0.9, suggesting that the
resultant catalyst maintains the mesoporous structure of the
carbon support (Figure 1D). The surface area and volume of
3
+
anchoring Fe ions.
2
À1
3
À1
Fe O /C are 459 m g and 0.39 cm g , respectively. The mor-
3
4
phology of Fe O /C was investigated with high-resolution
3
4
transmission electron microscopy. The TEM image of Fe O /C
3
4
(
Figure 2a) shows that the iron oxide is present as 30 nm
nanoparticles on the carbon support. The HR-TEM image of an
individual nanoparticle shows that the d-spacing of the lattice
fringes is approximately 0.24 nm, which can be ascribed to the
[
39,40]
(
311) plane of the Fe O crystals (Figure 2b).
Three well-re-
3
4
solved diffraction rings are detected in the selected-area elec-
tron diffraction (SAED) pattern, which can be indexed as the
(
311), (440), and (511) planes of Fe O4 (Figure 2c). The
3
energy-dispersive X-ray (EDX) image (Figure 2d) confirms the
presence of Fe, O, and C elements in the selected area.
The magnetic properties of Fe O /C was studied by using
3 4
Scheme 1. Schematic illustration of the preparation process of Fe O /C.
3
4
a vibrating sample magnetometer (VSM) at room temperature.
The applied magnetic field was varied from À20000 Oe to
20000 Oe (Figure 3). Fe O /C exhibits a superparamagnetic
Figure 1A shows the X-ray diffraction (XRD) pattern of
Fe O /C calcined at 4008C under an Ar flow. As a reference,
3
4
3
4
bulk Fe O4 was prepared by precipitating a stoichiometric
character with a negligible hysteresis loop. The remnant mag-
3
3
+
2+
[33,34]
À1
molar ratio of Fe and Fe ions with ammonia.
tively sharp diffraction peaks at 2q values of 30.2, 35.6, 43.3,
3.8, 57.3, and 62.98 can be observed in both Fe O /C and bulk
Six rela-
netization (M ) is very small (0.3 emug ), which means that
r
nearly no magnetization remained on Fe O /C when the exter-
3
4
5
nal magnetic field was removed. This property is favorable for
its application as a catalyst, as the magnetization cannot influ-
ence the dispersion of Fe O /C in the reaction system. The sat-
3
4
Fe O , which match well with the diffractions from the (220),
3
4
(
311), (400), (422), (511), and (440) lattice planes of cubic in-
3
4
[
35,36]
À1
verse spinel magnetite iron oxide (Fe O , JCPDS 65-3107).
uration magnetization of Fe O /C is 12.4 emug . This relatively
3 4
3
4
No peaks assigned to other phases of iron oxide (a-Fe O , for
high value can be attributed to the pure Fe O crystal phase
3 4
2
3
example) were found in the pattern. For Fe O /C, the broad
present in Fe O /C. Taking advantage of the relatively strong
3 4
3
4
peak centered at 24.58 can be assigned to the diffractions from
superparamagnetic property, Fe O /C can be easily separated
3 4
the (002) graphite planes of the carbon support (Figure S1a in
from reaction solution by using an external magnetic field
(Figure 3, inset).
the Supporting Information). It should be note that g-Fe O3
2
has the same XRD pattern as that of Fe O . To exclude the for-
The catalytic performance of Fe O /C was tested in the selec-
3
4
3
4
57
mation of any g-Fe O phase in the Fe O /C composite, Fe
tive oxidation of benzyl alcohol with air as the oxidant source
(Figure 4). Fe O /C exhibits 93.6% conversion of benzyl alcohol
2
3
3
4
Mçssbauer measurements were carried out (Figure 1B).
3
4
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
806
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
57
Figure 1. A) XRD patterns Fe
3 4 3 4 3 4 3 4 3 4 2
O /C and bulk Fe O . B) Fe Mçssbauer spectrum of Fe O /C. C) Raman spectra Fe O /C and bulk Fe O . D) N adsorption profiles
of Fe /C (inset: pore size distribution).
3
O
4
Table 1. The fine parameters of the fitted curves to the Mçssbauer spec-
[a]
trum of Fe
3 4
O /C.
Component
IS
mms
QS
[mms
Hf
[kOe]
WHWM
AREA
[%]
À1
À1
À1
[
]
]
[mms
]
A
B
Fe
0.27Æ0.01 0.04Æ0.04 494.4Æ0.9 0.32Æ0.01 39.7
0.55Æ0.01 À0.07Æ0.03 454.9Æ1 0.41Æ0.01 46.6
Fe
3
O
4
3
+
0.28Æ0.01 1.09Æ0.03
–
0.33Æ0.01 13.7
[
a] Where IS is the isomer shift relative to a-Fe at room temperature. QS is
57
the quadrupole shift. Hf is the magnetic hyperfine field at Fe nuclei.
WHWM is the half-maximum linewidth. AREA is the area of the spectrum
component.
after 8 h of reaction, which is comparable with that of support-
ed ultrafine iron oxide particles (<2 nm) reported previously
[
15]
and much higher than that of supported 5–10 nm a-Fe O .
2
3
The selectivity to benzaldehyde is nearly 100%. No byproducts
such as benzoic acid and/or benzophenone were detected by
GC-MS. Under the same reaction conditions, bulk Fe O as well
3
4
as the carbon support showed very poor catalytic activities
less than 5% after 8 h of reaction). This result excludes the
(
3 4
Figure 2. a) TEM, b) HR-TEM, c) SAED and d) EDX images of Fe O /C.
direct contribution of the carbon support to the title reaction.
A hot filtration test was carried out to confirm the heteroge-
neous nature of the Fe O /C catalyst. The results showed that
be easily recovered from the reaction system by magnetic sep-
aration. After five test cycles, the catalyst maintains relatively
3
4
no subsequent conversion of benzyl alcohol was detected in
the filtrate, confirming that Fe O /C really is a heterogeneous
high activity. XRD and N adsorption measurements were car-
3
4
2
catalyst (Figure S2 in the Supporting Information). In addition,
multiple aerobic oxidation cycles were also carried out to ex-
amine the recoverability of Fe O /C (Figure 5). The catalyst can
ried out to examine the structural properties of the catalyst
after five catalytic cycles (Figures S3 and S4 in the Supporting
Information). The results show that the crystallinity and meso-
3
4
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
807
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[a]
Table 2. Catalytic oxidation of aromatic alcohols over Fe
3
O
4
/C.
Entry
Substrate
Product
t
Yield
[%]
[h]
1
4
89.3
88.5
2
4
3
4
5
6
6
6
86.7
97.1
88.3
3 4
Figure 3. The magnetic hysteresis loop of Fe O /C.
6
7
12
12
81.9
66.3
[
b]
b]
8
9
12
12
65.5
41.6
[
[
(
a] Reaction conditions: substrate (1 mmol), catalyst (0.3 g), toluene
10 mL), 808C, air (1 bar). [b] Substrate (0.5 mmol), 1008C.
To illustrate the general applicability of Fe O /C, the catalyst
3
4
was extended to the aerobic oxidation of other alcohols under
relatively mild conditions (Table 2, entries 1–7). The results
show that Fe O /C also exhibits nice activity in aerobic oxida-
3 4 3 4
Figure 4. Catalytic performance of Fe O /C, bulk Fe O , and the carbon sup-
port in the aerobic oxidation of benzyl alcohol. Reaction conditions: benzyl
alcohol (1 mmol), catalyst (0.3 g), toluene (10 mL), 808C, air (1 bar).
3
4
tion of the benzyl alcohol derivatives with either electron-rich
or electron-poor substituents, in particular, for p-nitro benzyl
alcohol (89.3% yield after 4 h), 4-methoxy benzyl alcohol
(
97.1% yield after 6 h), and 4-hydroxy-3-methoxy benzyl alco-
hol (88.5% yield after 4 h). In addition, allylic and secondary al-
cohols such as trans-2-hexenol and sec-butyl alcohol can also
be oxidized into the corresponding aldehyde and ketone with
high selectivity over Fe O /C by increasing temperature and re-
3
4
action time (Table 2, entries 8 and 9). The above catalytic tests
clearly show that Fe O /C could serve the dual role of an effi-
3
4
cient catalyst and a magnetically recoverable entity in the
aerobic oxidation of alcohols.
To understand the correlation between the physicochemical
properties of Fe O /C and the catalytic activity, temperature-
3
4
programmed reduction (H -TPR, Figure 6) and X-ray photon
2
spectroscopy (XPS, Figure 7) were carried out. As a reference,
bulk a-Fe O3 was prepared by a conventional precipitation
Figure 5. Reusability of Fe
Reaction conditions: benzyl alcohol (1 mmol), catalyst (0.3 g), toluene
10 mL), 808C, 6 h, air (1 bar). The reaction time for the latter two cycles is
3
O
4
/C in the aerobic oxidation of benzyl alcohol.
2
method and identified by XRD patterns (Figure S5 in the Sup-
(
porting Information). Previously, we have reported that bulk a-
8
1
h and 10 h and the recovered catalysts were treated under Ar at 4008C for
h before another cycle to remove absorbed species.
[16]
Fe O is almost inactive in the aerobic oxidation of alcohols.
2
3
H -TPR shows that three reduction peaks are present for the
2
samples of bulk a-Fe O . The first narrow peak (centered at
2
3
porous structure underwent no obvious changes during reac-
tion. This result suggests that Fe O /C possesses high stability
4308C) can be ascribed to the reduction of a-Fe O to Fe O .
2 3 3 4
The two overlapped peaks in the range 450–8008C can be as-
3
4
[
41,42]
under these reaction conditions.
cribed to further reduction to FeO/Fe species.
Compared
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
808
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Fe O /C, only one broad peak is present in the range 370–
3
4
6
008C, and the center of this peak is clearly lower than that of
bulk Fe O . All the results above show that the Fe O in Fe O /
3
4
3
4
3
4
C is more reducible than that in bulk Fe O .
3
4
XPS was carried out to investigate the chemical states of Fe
and O on the surface of Fe O /C and bulk Fe O . As shown in
3
4
3
4
Figure 7A, the Fe2p core level is split into Fe2p1/2 and Fe2p3/2
in the range of the Fe2p binding energies and the peaks are
broad for both Fe O /C and bulk Fe O owing to the presence
3
4
3
4
2
+
2+
[39]
of Fe (2p3/2) and Fe (2p ). The center of the peaks for
1/2
Fe O /C shift to higher binding energies compared with those
3
4
for bulk Fe O , suggesting relative high Fe valence in Fe O /C.
3
4
3
4
The deconvolution of the O1s XPS spectra according to the
[
43–45]
binding energy of O1s has been reported in the literature.
For bulk Fe O , the O1s spectra can be deconvoluted into two
Figure 6. H
the normalized H
the content of Fe
2
-TPR profiles of Fe
3
O
4
/C, bulk Fe
3
O
4
, and a-Fe
2
O
3
. Inset shows
based on
2
consumption peaks of Fe
3
O
4
/C and bulk Fe O
3 4
3
4
3
4
O .
peaks (Figure 7B). Peak I, in the range 529.5–530.5 eV, can be
2À
assigned to “O ” ions in the lattices of iron oxide. Peak II, cen-
tered at about 532 eV, is related to the oxygen in the surface
OH groups or oxygen vacancies of Fe O . The area ratio of
3
4
Peak II/Peak I is 0.41 for bulk Fe O . For Fe O /C, the O1s spec-
3
4
3
4
tra could be deconvoluted into three peaks. The new peak
(
Peak III) at 533 eV can be attributed to the oxygen functional
[
46,47]
groups that remain on the surface of the catalyst.
The area
ratio of Peak II/Peak I is 2.43 for Fe O /C, which is much higher
3
4
than that of bulk Fe O . This result shows that Fe O in Fe O /C
3
4
3
4
3
4
possesses a large amount surface OH groups or oxygen vacan-
cies. This unique property probably originates from the small
particle size effect of Fe O and the interaction between the
3
4
[
48–50]
particle and carbon support.
the high reducibility of Fe O /C.
It should be responsible for
3
4
Recently, the correlation between the reducibility of iron
oxide and its catalytic activity in oxidation reactions has been
[
50–52]
widely investigated.
It has been accepted that the reduci-
bility of iron oxide catalysts plays an important role in deter-
mining their catalytic performance. Fe O4 itself has a high
3
oxygen storage capacity and flexible transitions between the
3
+
2+
Fe and Fe oxidation states. In this case, the small particle
size and the interaction between the Fe O4 particles and
3
carbon support endow the catalyst with higher reducibility.
Thus, the oxidation state of these active sites is more easily
changed. This might be responsible for the high activity of
Fe O /C in the aerobic oxidation of alcohols.
3
4
3 4 3 4
Figure 7. A) Fe2p and B) O1s XPS spectra of Fe O /C and bulk Fe O .
Conclusions
3
+
2+
with a-Fe O , the first peak (reduction of Fe
to Fe ) of
A composite of Fe O nanoparticles anchored on carbon has
2
3
3
4
Fe O is located at a much lower temperature, suggesting the
been prepared by an ammonia-assisted precipitation method.
The Fe O /C composite serves the dual role of a catalyst and
3
4
higher reducibility of Fe O . For Fe O /C, the center of the first
3
4
3
4
3
4
peak is the same as that of bulk Fe O . Apart from the reduc-
a magnetically recoverable entity in the aerobic oxidation of al-
cohols. The phase and particle size of iron oxide were con-
trolled well by introducing the carbon support, ammonia, and
ethanol solution. Fe O nanoparticles in Fe O /C have relatively
3
4
tion temperature, the discussion of redox properties of the
iron oxides as indicated by the TPR profiles should also consid-
er the amount of Fe O in the sample of Fe O /C. As shown in
3
4
3
4
3
4
3
4
the Experimental Section, the amount of Fe O is 20 wt% in
high reducibility, and their oxidation state are easily changed.
3
4
the sample of Fe O /C. If normalized to the contents of Fe O ,
These unique intrinsic properties of the Fe O4 nanoparticles
3
4
3
4
3
the reduction peak centered at 3298C for Fe O /C is much
are probably responsible for the high activity of Fe O /C in the
3
4
3
4
larger than that of bulk Fe O (Figure 6, inset). In addition, for
aerobic oxidation of alcohols.
3
4
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
809
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
formed on a Tianjin Xianquan TP-5079 adsorption analyzer. Before
detection, the catalysts (30 mg) were firstly treated in Ar (99.99%)
flow at 4008C for 10 min.
Experimental Section
Synthetic procedures
All the chemical reagents were obtained from commercial sources
and used without further purification.
General procedure for catalytic reaction
The mesoporous carbon support was prepared by using a modified
[28,29]
The aerobic oxidation of benzyl alcohols was carried out in
a 50 mL two-neck round-bottom flask at atmospheric pressure.
Typically, a mixture of alcohol (1 mmol), catalyst (0.3 g), and tolu-
ene (10 mL) was added into the flask and vigorously stirred at
hard-template method.
Typically, phosphoric acid was dropped
into an aqueous solution containing aluminum nitrate nonahy-
drate, citric acid, and sucrose. Then, a dilute aqueous ammonia
(
5
10%) solution was used to adjust the pH value of the solution to
.1. After that, the mixture was heated at 808C until water and
8
08C under an air balloon for 8 h. Aliquots of the products were
taken with a sampling pipe and analyzed with a gas chromato-
graph equipped with a HP-5 column and FID detector. The product
mixtures were further identified by GC-MS. The hot filtration test
was carried out as follows: after 2 h of reaction, the solid catalysts
were separated with a Buchner funnel. The mixture of the filtrate
was put into the reactor and continuously reacted under the same
conditions (808C, air 1 atm) without the solid catalyst. In the recy-
cling experiment, the solid catalyst was separated from the reac-
tant by an external magnet after reaction and further treated at
other volatiles were removed to obtain the as-synthesized precur-
sor. The solid was first pretreated at 3008C in muffle for 10 min,
and then calcined at 8008C under an argon atmosphere for 6 h. Fi-
nally, the composite was treated with 4m HCl or HNO at 608C for
3
6
h to remove the AlPO component. The resulting products were
filtered, washed with deionized water, and dried at 1008C over-
night.
Fe O /C was synthesized at ambient atmospheric pressure by
3
4
a modified ammonia-assisted precipitation method. In a typical
4
008C for 1 h before the next test cycle.
procedure, Fe(NO ) ·9H O (0.72 g) was dissolved in an ethanol/
3
3
2
3
+
water solution (80.0 mL, 1:1) to get a homogeneous Fe solution
before the carbon support (0.5 g) was added. Then, a dilute alkali
aqueous solution (10% ammonia) was used to slowly adjust the
pH value of the solution to 8.0 and the mixture was further stirred
for about 3 h. The resulting mixture were filtered, washed with de-
ionized water, and dried at 808C in the oven overnight. To get the
iron oxide into the desired crystalline structure, the obtained
sample was further pyrolyzed under an argon atmosphere at
Acknowledgements
This work was financially supported by the National Natural Sci-
ence Foundation of China (21003059, 21473073), the Develop-
ment Project of Science and Technology of Jilin Province
(20130101014JC), the Fundamental Research Funds for the Cen-
tral Universities and the Open Project of State Key Laboratory of
Inorganic Synthesis and Preparative Chemistry. Dr Xiang Wang
gratefully acknowledges the US Department of Energy (DOE),
Office of Science, Office of Basic Energy Sciences, Chemical Scien-
ces, Geosciences, and Biosciences Division for support of this
work.
4
008C for 4 h. The weight of Fe O is about 20 wt% in the resul-
3 4
tant catalyst.
For comparison, bulk a-Fe O and Fe O were prepared. a-Fe O
3
2
3
3
4
2
was obtained with a conventional precipitation method using
Fe(NO) ·9H O as the iron source and ammonia as the precipitation
3
2
agent (pH=9). The resultant material was calcined at 4008C for
2
+
4
h. The bulk Fe O was prepared by co-precipitation of Fe and
3 4
3
+
[30,31]
Fe with an ammonia solution.
Keywords: alcohol oxidation
· iron catalysis · magnetic
separation · mesoporous carbon · nanoparticles
Catalyst characterization
[
1] N. D. Spencer, R. C. Schoonmaker, G. A. Somorjai, Nature 1981, 294,
The crystal structures were recorded by using a Rigaku Powder X-
643–644.
ray diffractometer (XRD) equipped with a CuK radiation source
[2] R. V. Jagadeesh, A.-R. Henrik Junge, M.-M. Pohl, J. Radnik, J. Rabeah, H.
Huan, V. Schünemann, A. Brückner, M. Beller, Science 2013, 342, 1073–
1076.
a
5
7
(
l=1.5418 Š). Fe Mçssbauer spectra were recorded by using
5
7
a OIMS-500 Mçssbauer spectrum instrument. Co(Pd) was used as
the g-ray radioactive source and the velocity was calibrated against
a standard a-iron foil. Raman spectra were recorded on a Bruker
RFS 100 Raman spectrometer with an argon laser (532 nm) as the
excitation source. The morphology, energy-dispersive absorption
X-ray spectra (EDX) and selected-area electron diffraction (SAED)
images were determined by using a FEI Tecnai F20 EM transmission
electron microscopy (TEM) operating at an accelerating voltage of
[
3] C. Yang, H. Zhao, Y. Hou, D. Ma, J. Am. Chem. Soc. 2012, 134, 15814–
5821.
4] K. M. Driller, H. Klein, R. Jackstell, M. Beller, Angew. Chem. Int. Ed. 2009,
8, 6041–6044; Angew. Chem. 2009, 121, 6157–6160.
1
[
4
[
[
[
5] Y. Zou, J. Kan, Y. Wang, J. Phys. Chem. C 2011, 115, 20747–20753.
6] S. A. Cotton, Annu. Rep. Prog. Chem. Sect. A 2005, 101, 208–230.
7] B. Plietker, Iron Catalysis in Organic Chemistry, Wiley-VCH, Weinheim,
2008.
2
00 kV. XPS was performed on a Thermo ESCA LAB 250 system
[8] G. Bai, X. Lan, X. Liu, C. Liu, L. Shi, Q. Chen, G. Chen, Green Chem. 2014,
16, 3160–3168.
with MgK source (1254.6 eV). N adsorption–desorption isotherms
a
2
[
9] C. Bolm, J. Legros, J. L. Paih, L. Zani, Chem. Rev. 2004, 104, 6217–6254.
10] S. Enthaler, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2008, 47, 3317–
321; Angew. Chem. 2008, 120, 3363–3367.
11] B. Chen, J. Li, W. Dai, L. Y. Wang, S. Gao, Green Chem. 2014, 16, 3328–
334.
were measured at 77 K by using a Micromeritics ASAP 2010N ana-
lyzer. Samples were degassed at 2008C for 20 h before measure-
ments. Specific surface areas were calculated by using the Brun-
auer–Emmett–Teller (BET) model. Pore size distributions were eval-
uated from desorption branches of the nitrogen isotherms by
using the Barret–Joyner–Halenda (BJH) model. Magnetic measure-
ments were performed on a SQUID-VSM magnetometer at room
temperature. Temperature-programmed reduction (TPR) was per-
[
[
[
3
3
12] F. Shi, M. K. Tse, M. M. Pohl, J. Radnik, A. Brückner, S. M. Zhang, M. Beller,
J. Mol. Catal. A 2008, 292, 28–35.
[13] B. Moens, H. D. Winne, S. Corthals, H. Poelman, R. D. Gryse, V. Meynen,
P. Cool, B. F. Sels, P. A. Jacobs, J. Catal. 2007, 247, 86–100.
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
810
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
[
[
[
14] M. Qing, Y. Yang, B. S. Wu, J. Xu, C. H. Zhang, P. Gao, Y. W. Li, J. Catal.
011, 279, 111 –122.
15] L. Geng, X. Zhang, W. Zhang, M. Jia, G. Liu, Chem. Commun. 2014, 50,
965–2967.
16] L. Geng, M. Zhang, W. Zhang, M. Jia, W. Yan, G. Liu, Catal. Sci. Technol.
015, 5, 3097–3102.
17] R. Hudson, Y. Feng, R. S. Varma, A. Moores, Green Chem. 2014, 16,
493–4505.
[35] E. Kang, Y. S. Jung, A. S. Cavanagh, G.-H. Kim, S. M. George, A. C. Dillon,
J. K. Kim, J. Lee, Adv. Funct. Mater. 2011, 21, 2430–2438.
[36] C. Ban, Z. Wu, D. T. Gillaspie, L. Chen, Y. Yan, J. L. Blackburn, A. C. Dillon,
Adv. Mater. 2010, 22, E145–E149.
[37] X. Zhang, Y. Niu, X. Meng, Y. Li, J. Zhao, CrystEngComm 2013, 15, 8166–
8172.
2
2
2
[38] J. Tu, M. Ding, Q. Zhang, Y. Zhang, C. Wang, T. Wang, L. Ma, X. Li, Chem-
CatChem 2015, 7, 2323–2327.
4
[
[
18] H. Li, J. Liao, X. Zhang, J. Mater. Chem. A 2014, 2, 17530–17535.
19] G. Yan, Y. Jiang, C. Kuang, S. Wang, H. Liu, Y. Zhang, J. Wang, Chem.
Commun. 2010, 46, 3170–3172.
[39] J. Lu, X. Jiao, D. Chen, W. Li, J. Phys. Chem. C 2009, 113, 4012–4017.
[40] H. Yu, Y. Li, X. Li, L. Fan, S. Yang, J. Mater. Chem. A 2014, 2, 15763–
15767.
[
[
[
[
20] H. Firouzabadi, N. Iranpoor, M. Gholinejad, J. Hoseini, Adv. Synth. Catal.
[41] X. Mou, B. Zhang, Y. Li, L. Yao, X. Wei, D. Su, W. Shen, Angew. Chem. Int.
Ed. 2012, 51, 2989–2993; Angew. Chem. 2012, 124, 3044–3048.
[42] B. Qiao, A. Wang, L. Li, Q. Lin, H. Wei, J. Liu, T. Zhang, ACS Catal. 2014,
4, 2113–2117.
[43] M. D. Sµnchez, P. Chen, T. Reinecke, M. Muhler, W. Xia, ChemCatChem
2012, 4, 1997–2004.
2
011, 353, 125–132.
21] H. Mahmoudi, A. A. Jafari, S. Saeedi, H. Firouzabadi, RSC Adv. 2015, 5,
023–3030.
22] S. Yang, C. Gao, Y. Sun, P. Huang, F. Wei, W. Song, Angew. Chem. Int. Ed.
015, 54, 2661–2664; Angew. Chem. 2015, 127, 2699–2702.
3
2
23] K. C. F. Leung, S. Xuan, X. Zhu, D. Wang, C.-P. Chak, S.-F. Lee, W. K. W.
Hob, B. C. T. Chung, Chem. Soc. Rev. 2012, 41, 1911–1928.
[44] Q. Zhuo, J. Gao, M. Peng, L. Bai, J. Deng, Y. Xia, Y. Ma, J. Zhong, X. Sun,
Carbon 2013, 52, 559–564.
[
[
24] R. L. Oliveira, P. K. Kiyohara, L. M. Rossi, Green Chem. 2010, 12, 144–149.
25] K. Park, G. Liang, X. Ji, Z.-P. Luo, C. Li, M. C. Croft, J. T. Markert, J. Phys.
Chem. C 2007, 111, 18512–18519.
[45] G. Chen, Y. Zhao, G. Fu, P. N. Duchesne, L. Gu, Y. Zheng, X. Weng, M.
Chen, P. Zhang, C.-W. Pao, J.-F. Lee, N. Zheng, Science 2014, 344, 495–
499.
[
[
26] Y. Pan, X. Du, F. Zhao, B. Xu, Chem. Soc. Rev. 2012, 41, 2912–2942.
27] H. Firouzabadi, N. Iranpoor, M. Gholinejad, S. Akbari, N. Jeddi, RSC Adv.
[46] L. Geng, S. Wu, Y. Zou, M. Jia, W. Zhang, W. Yan, G. Liu, J. Colloid Inter-
face Sci. 2014, 421, 71–77.
2
014, 4, 17060–17070.
[47] M. Zhang, W. Zhu, G. Liu, X. Zhang, Y. Zu, W. Zhang, W. Yan, M. Jia, Chin.
J. Catal. 2012, 33, 465–472.
[
[
28] F. Shi, M. K. Tse, M.-M. Pohl, A. Brückner, S. Zhang, M. Beller, Angew.
Chem. Int. Ed. 2007, 46, 8866–8868; Angew. Chem. 2007, 119, 9022–
[48] M. Cargnello, V. V. T. Doan-Nguyen, T. R. Gordon, R. E. Diaz, E. A. Stach,
R. J. Gorte, P. Fornasiero, C. B. Murray, Science 2013, 341, 771–773.
[49] Q. Fu, W. Li, Y. Yao, H. Liu, H. Su, D. Ma, X. Gu, L. Chen, Z. Wang, H.
Zhang, B. Wang, X. Bao, Science 2010, 328, 1141–1143.
[50] K. Zhao, H. Tang, B. Qiao, L. Li, J. Wang, ACS Catal. 2015, 5, 3528–3539.
[51] B. Qiao, A. Wang, X. Yang, L. F. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li, T.
Zhang, Nat. Chem. 2011, 3, 634–641.
9
024.
29] Y. Gao, D. Ma, G. Hu, P. Zhai, X. Bao, B. Zhu, B. Zhang, D. Su, Angew.
Chem. Int. Ed. 2011, 50, 10236–10240; Angew. Chem. 2011, 123, 10419–
1
0423.
[
[
30] H. Wei, E. Wang, Chem. Soc. Rev. 2013, 42, 6060–6093.
31] G. Liu, Y. Liu, Z. Wang, X. Liao, S. Wu, W. Zhang, M. Jia, Microporous Mes-
oporous Mater. 2008, 116, 439–444.
32] G. Liu, Z. Wang, M. Jia, X. Zou, X. Zhu, W. Zhang, D. Jiang, J. Phys. Chem.
B 2006, 110, 16953–16960.
[52] L. Li, A. Wang, B. Qiao, J. Lin, Y. Huang, X. Wang, T. Zhang, J. Catal.
2013, 299, 90–100.
[
[
[
33] N. Anand, K. H. P. Reddy, T. Satyanarayana, K. S. R. Rao, D. R. Burri, Catal.
Sci. Technol. 2012, 2, 570–574.
34] F. Jiao, J.-C. Jumas, M. Womes, A. V. Chadwick, A. Harrison, P. G. Bruce, J.
Am. Chem. Soc. 2006, 128, 12905–12909.
Received: October 20, 2015
Published online on January 13, 2016
ChemCatChem 2016, 8, 805 – 811
www.chemcatchem.org
811
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim